The Advanced ML Engineering – Industrial AI Development Group at Seagate develops transparent, reliable machine learning capabilities to strengthen semiconductor manufacturing. The role aims to improve model interpretability, identify key drivers, and provide actionable insights for engineers and operators to enhance decision-making and process understanding.
Requirements
- Strong experience with Python, LightGBM, and common ML libraries (scikit-learn, pandas, numpy).
- Solid understanding of model interpretability, including SHAP, feature attribution, model diagnostics, and visualization techniques.
- Familiarity with causal inference concepts and their use in understanding model behavior and real-world processes.
- Demonstrated interest in interpretable ML, LightGBM modeling, causal analysis, and decision-support systems for complex real-world processes.
- Experience building explainable AI workflows for real-world decision-support or quality engineering environments.
- Exposure to causal discovery (PC, GES, LiNGAM), counterfactual reasoning, or causal feature importance.
- Hands-on experience with anomaly detection, reliability prediction, or multivariate sensor analytics.
Responsibilities
- Build and refine interpretation tools for LightGBM models, including per-sample explanations and feature contribution summaries.
- Analyze model outputs to identify key drivers, inconsistencies, and signals that can guide process investigation.
- Apply causal reasoning—such as distinguishing correlated vs. influential features—to improve diagnosis and deepen understanding of model decisions.
- Create visualizations and dashboards that make model behavior understandable to engineers and operators.
- Document your methods and communicate insights to cross-functional teams to support deployment and continuous improvement.
Other
- Ability to translate complex technical insights into clear, domain-relevant narratives for manufacturing engineers.
- Demonstrated curiosity, strong analytical thinking, and ability to drive investigations independently.
- Effective communication and collaboration skills across technical and operational teams.
- Currently pursuing a Master’s or PhD in Computer Science, Data Science, Electrical Engineering, Industrial Engineering, or a related field, and returning for classes in Fall 2026.
- Familiarity with high-volume manufacturing, statistical process control, yield improvement, or root-cause analysis.
